The proposed study is designed to identify etiologic factors for preterm premature rupture of the membranes (preterm PROM) in which preterm delivery is precipitated by rupture of the chorioamnionic membranes. Exposures which decrease the strength of the membranes are postulated to predispose to rupture, focusing on genital tract infections, vitamin C, and tobacco smoke. This nested case-control study will be conducted within the cohort of 8,000 women who obtain prenatal care at three high- volume clinics affiliated with the University of North Carolina Hospitals and two clinics affiliated with Wake Medical Center Obstetrical Teaching Services (approximately half black and half white) over the period of January 1, 1995 to December 31, 1997. All 8,000 patients will have specimens collected at 22-28 weeks of gestation and stored for future assessment of genital tract infection, plasma and leukocyte levels of vitamin C, urinary cotinine, and urinary drug metabolites. The 250 women who develop preterm PROM, a sample of 250 women who deliver early due to preterm labor, and 500 controls will be recruited, with an expected 80% response to yield 200 preterm PROM cases, 200 preterm labor cases, and 400 controls. Preterm PROM and preterm labor cases will be identified in the hospital and controls will selected from prenatal clinics, with the same gestational age distribution as preterm PROM cases. For selected cases and controls, we will assay the stored genital tract specimens to identify bacterial vaginosis, chlamydia, and trichomonas infections, assay the blood sample for vitamin C, and assay the urine specimen for cotinine and illicit drugs. All cohort members will be interviewed to obtain information on diet, tobacco, drug, and alcohol use, sexual activity late in pregnancy, reproductive history, and other potential risk factors. The analyses will focus on the impact of specific genital tract infections, vitamin C, and smoking on the risk of preterm PROM by contrasting preterm PROM cases to controls. Comparison of the preterm labor cases to controls will suggest whether preterm PROM and preterm labor are etiologically distinct entities based on shared or distinct risk factors. This study will be the first large- scale evaluation of preterm PROM that simultaneously addresses genital tract infection, nutrition, and tobacco with biological assessments as well as detailed self-report in a large bi-racial population. Although all are strongly suspected of influencing risk of preterm delivery, this study should provide evidence relevant to screening practice for infection and vitamin use during pregnancy.